The overall objective of this project is to continue evaluation of the promising hypoxia imaging agent, [F-18] fluoromisonidazole. The aims are to determine how specifically this radiopharmaceutical localizes in hypoxic human tumors of different histologies; whether this localization predicts a poor response to photon radiotherapy or can be used to assess tumor reoxygenation; and how best to model the dynamic time activity data to assess the extent of radiobiologic hypoxia in the imaged tumor volume. Measurable human tumors. particularly locally advanced head and neck, prostate, and non-small cell lung cancers that are to be treated primarily with irradiation will be imaged with [F-18]FMISO and PET. FMISO retention in these neoplasms will be correlated with response to x-ray or neutron radiotherapy, with the latter therapy presumed to be relatively little influenced by tumor hypoxia. Decreases in FMISO uptake during fractionated therapy will be measured as an indicator of tumor reoxygenation. Correlative studies of [H-3]FMISO uptake and retention in human tumor cells in vitro and in rats bearing multiple subcutaneous tumors will provide data to help validate the FMISO imaging and modeling methods, bridging between models previously developed for multicellular spheroids and those models being developed to quantify hypoxic fraction in human cancers imaged with PET.